exa-search

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses curl and jq to interact with the Exa API and process JSON responses. This is standard behavior for a CLI-based API integration.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill manages an API key stored at ~/.config/exa-search/config.json. It provides instructions for users to safely create this file and populate it with their own credentials. There are no patterns indicating the key is sent to any unauthorized third-party domains; all traffic is directed to api.exa.ai.
  • [INDIRECT_PROMPT_INJECTION]: As a search skill, it ingests untrusted data from the web (via /search and /contents). While it lacks explicit boundary markers in its output examples, it is a single-purpose tool where the risk of downstream exploitation is minimal compared to the primary function of web retrieval.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 05:15 PM